The aims in this study will (1) develop candidate neurocognitive and neuroimaging endophenotypes for bipolar I disorder (BPI), (2) examine the association of history of psychosis and these brain-behavior markers in BPI patients, and (3) determine if markers are sensitive to liability for psychosis. Thus, the project has two overlapping goals: the development of candidate endophenotypes for BPI broadly and the identification of candidate endophenotypes for psychosis. The ultimate promise of this research is to develop markers that will characterize the biological mechanisms of BPI and facilitate the discovery of genes that predispose the illness. We believe that a comprehensive assessment of neuropsychological functioning and brain structure and function in sibling pairs discordant for bipolar disorder and stratified for psychosis history is a strategically strong first step towards reaching these goals. To this end, we will perform anatomic and functional neuroimaging and conduct neuropsychological examinations on 130 euthymic patients with BPI (65 with history of psychosis), 130 of their unaffected same-sex siblings and 65 unrelated comparison subjects. Markers found to be aberrant in both affected individuals (regardless of history of psychosis) and in their unaffected relatives will be considered candidate endophenotypes for BPI (Aim 1). Neuropsychological and neuroimaging measures that distinguish between BPI patients with and without history of psychosis (Aim 2) and their siblings (Aim 3) will be considered potential endophenotypes for psychosis. Bipolar I disorder represents a significant economic burden and is associated with substantial morbidity and mortality rates. Although it is well established that BPI is substantially heritable, the molecular genetic basis for this illness remains elusive, potentially because of illness complexity, heterogeneity of disease expression, and comorbidity with other disorders that may distort clinical presentation. In the face of evidence that genes predisposing to BPI may be transmitted without expression of the clinical phenotype, interest has arisen in developing endophenotypes for the illness, indicators of processes mediating between genotype and phenotype. Given the high rates of psychosis in BPI and that history of psychosis may alter brain structure and function, we believe that elucidating neurocognitive and neuroimaging endophenotypes for the disorder must account for the potential impact of hallucinations and delusions on these markers. This research will established biomarkers that could improve the identification and treatment of BPI patients and, potentially, patients with psychotic disorders, independent of their formal diagnosis. PUBLIC HEALTH RELEVANCE: Bipolar I disorder is a major public health burden whose biology is still largely unknown. Through the development of brain-behavior markers sensitive to genetic liability for bipolar disorder, the proposed research should significantly aid the discovery of genes that predispose the illness and facilitate the identification of the biological determinants of bipolar disorder. This, in turn, should lead to improved characterization and treatment of bipolar disorder.